Tide dependent seasonal changes in water quality and assimilative capacity of anthropogenic influenced Mormugao harbour water

Indian Journal of Geo-Marine Sciences Vol. 41(4), August 2012, pp. 314-330

Tide dependent seasonal changes in water quality and assimilative capacity of

anthropogenically influenced Mormugao harbour water

P. V. Shirodkar, M. Deepthi, P. Vethamony, Analia M. Mesquita, U. K. Pradhan,

M. T. Babu, X.N.Verlecar & Sonali R. Haldankar

National Institute of Oceanography, Council of Scientific and Industrial Research, Dona Paula, Goa – 403 004, India [E.mail: [email protected]]

Received 02 February 2011; revised 27 July 2011

Water quality data from Mormugao harbour area at the mouth of Zuari estuary in Goa have been obtained over two tidal cycles monthly for the year 2003 to 2004. R-mode factor analyses of the data indicated strong positive loadings of coliforms, nitrogenous substances, PHc and some heavy metals during post-monsoon, suggesting their dominance and significant increase relative to other seasons. Water quality index (WQI) studies showed overall index of pollution (OIP)

values of 1.41 to 3.52, categorizing the harbour water as slightly polluted during pre-monsoon. During post-monsoon, the OIP values of 1.75 to 5.4 categorized the harbour water as polluted water, while it was acceptable quality water during the

monsoon, with OIP values of 1.13 to 1.79. The higher chlorophyll a (1.4-19 mgm–3µg/l) and primary productivity (4.8-14.8

mg/l) of the monsoon was followed by that of the post-monsoon season, with the least during pre-monsoon. Modeling studies using Ecolab module of MIKE 21 suggested slightly higher assimilative capacity for harbour water during the pre-monsoon relative to post-monsoon, which showed a reduction in assimilative capacity.

[Keywords: Mormugao harbour, Physico-chemical data, Factor analyses, Water quality index , Numerical modeling].

Introduction

Increasing population, urbanisation and

industrialisation pump several tones of wastes into the coastal water via the riverine/estuarine systems as

well as through direct discharges1,. This has

contaminated the estuarine and coastal waters2,3

, at many locations along the Indian coat and affected

their flora and fauna4-6

. Zuari river in Goa, along the

west coast of India, is one such river, where

anthropogenic wastes are gradually contaminating its estuarine environment. The ability of a water body to

assimilate anthropogenic waste varies from coastal to

estuarine to the freshwater environment7. The semi-

enclosed coastal waters generally accumulate more

anthropogenic waste because of their lower ability to

flush the contaminants out to the sea as compared to those with open boundaries to the sea

8. Larger the

assimilative capacity of a water body, lesser is the

change in its water quality. Understanding the

changes in water quality and assimilative capacity of a water body is one of the important aspects of marine

pollution studies. Thomann and Mueller, (1987)9

indicated that the DO, BOD and nutrients, which help in assessing the quality of a water body, also assist in

understanding the waste assimilative capacity of

water. In recent times, the numerical modeling

(Ecolab module) is being largely used for

understanding the waste assimilative capacity of water and the water quality deterioration from sewage

and industrial discharges10-16

.

Mormugao port located in the estuarine region of zuari has large activities. There are two Sewage

Treatment Plants (STPs) located on its either side, one

at Mormugao Headland, near the mouth of Zuari, and

the other one by the side of the mouth of the estuary at Baina . Along the southern bank of Zuari in line

with Mormugao port, there are shipbuilding

industries, yards, workshops, etc. and various other anthropogenic setups. The anthropogenic wastes

generated in this regime include washings from ships

cleaning, scraped paints and oil, metal wastes and various other domestic wastes which are ultimately

discharged into the Zuari estuary.

One of the major concerns of Mormugao harbour

region is the contamination resulting from anthropogenic activities including the waste

discharges from Mormugao Sewage Treatment Plant

(STPs) due to the inward and outward movement of contaminants with respect to tides in the harbour

region of Zuari estuary and their impact on water

SHIRODKAR et al : TIDE DEPENDENT SEASONAL CHANGES IN WATER QUALITY

315

quality and productivity. This has not been explored

properly. In this study, an attempt has been made to

understand how the dominant anthropogenic contaminants affect water quality, the assimilative

capacity and productivity of Mormugao harbour water

over different seasons, using multivariate statistics,

water quality index (WQI) and Ecolab module. The role of hydrodynamics in maintaining the water

quality of Mormugao harbour water has been studied

using the Hydrodynamic module.

Material and Methods

The present study area is Mormugao harb

our region located near the estuarine mouth of Zuari

(15o

22’ N to 15o 28’ N latitude and 73

o 44’ E to 73

o

51’ E longitude) (Figure 1). Maximum depth of this region is 20 m and average depth is 3 m. The STP’s

are located near the mouth of the estuary of which one

is outside the mouth of the estuary towards its south (Figure 2). Surface and bottom water samples were

regularly collected every month during high tide and

low tide from eight selected stations in Mormugao

harbor (Figure 1) from September 2003 to April 2004. These stations were selected on a total of 3 transects

(two vertical and one horizontal) in such a way that

they cover sufficiently a larger area of Mormugao harbour. Sampling periods earmarked for the study

comprised of 3 seasons, starting from monsoon of

2003 (June to September 2003); post-monsoon (October 2003 to January 2004) and Pre-monsoon

(February to May 2004). The data collected during

September 2003 has been considered as the one

representing here to the monsoon season. Besides this, 7 more stations (LC1 to LC7) with 4 stations within

and around the Mormugao harbour mouth and 3

stations south of the harbour mouth towards Baina as shown in Figure 2 were examined as a part of the

study. . During sample collection, the DO was first

fixed onboard using Winkler’s reagents, while BOD was analysed after 5 days of incubation at the shore

laboratory; Temperature and pH were measured on

board using a thermometer and Eutech pH meter.

The nutrients, chlorophyll a (Chl.a), petroleum hydrocarbons (PHc) and phenol were analysed at the

shore laboratory as per the standard methods17-19

.

The trace metals (lead, cadmium and mercury) in water samples were pre-concentrated by chelating

with ammonium pyrrolidine dithiocarbamate (APDC),

then extracted with methyl isobutyl ketone (MIBK)

and analysed by AAS. Total viable count (TVC), total

coliforms (TC) and total vibrios (TV) were analysed

from each of the samples as per the procedure given

by APHA, 199218

. Tides and currents were studied from Karwar,

Vengurla and Mormugao area and also simulated.

Bathymetry data required for simulation were

generated by digitizing the Naval Hydrographic charts (Chart No. 2022 for Mormugao; Chart No. 2043 for

Vengurla – Mormugao and Chart No. 215 for Karwar

– Mormugao) of the region, while the wind data was collected from the Autonomous Weather Station

(AWS) installed at the National Institute of

Oceanography, Dona Paula, Goa at a height of 45 m

and the data was reduced to 10 m level and used for simulation. For simulation of water level, the currents

and water quality parameters of the harbour region

were taken into account for running the Hydrodynamic and Ecolab modules of MIKE 21

Fig. 1—Map showing location of sampling stations in Mormugao harbour region of Zuari estuary.

Fig. 2—Map showing the Mormugao STP and Baina STP in Zuari estuary. The arrows indicate the flow pattern within Zuari estuary.

INDIAN J. MAR. SCI., VOL. 41, NO. 4, AUGUST 2012

316

numerical modeling software20-24

. The hydrodynamic

module calculated the hydrodynamic behavior of

water in response to a variety of forcing functions, while the Ecolab module described the chemical,

biological and ecological responses and interactions

between the stated variables. In Ecolab module, the

depletion in DO was considered related to the BOD content of water.

Results

Among the physico-chemical and biological variables measured from Mormugao harbour water,

the significant variations were shown by the dissolved

oxygen (DO), biochemical oxygen demand (BOD), phosphate, nitrate, silicates, petroleum hydrocarbons,

trace metals (Pb, Cd and Hg), chlorophyll a, primary

productivity, total viable count (TVC), total coliforms

(TC) and total vibrios (TV). Their ranges of variation over tidal changes during each season are given in

Table 1.

DO show a wide range of variation from 1.5 - 6.6 mgl

-1, with low values at some stations and high

values (> 6 mgl-1

) at others during monsoon. During

pre- and post-monsoon, normal values of > 3 mgl-1

prevailed. BOD showed a variation from 0.2 to

4.6 mgl-1

(av.1.2 mgl-1

), with higher values of

3 – 4.6 mgl-1

at some stations during the pre-monsoon

and lowest values of 0 to 2.0 mgl-1

(av. 0.6 mgl-1

) during the post-monsoon. Monsoon month showed

moderate values. Phosphate showed a large

variation, with significantly higher values during the

post-monsoon (0.1 – 13.8 µmoll-1

) and monsoon

(1.2 – 4.8 µmoll-1

) relative to pre-monsoon.

Considerable nitrate in the range of 0 – 5 µmoll-1

was observed during the pre-monsoon and post-monsoon

season (0 – 6.3 µmoll-1

), relative to monsoon. Since silicates are normally brought by the riverine water, the monsoon season showed high silicates (12 – 30

µmoll-1

; av. 21µmoll-1

) being added to the estuarine water by the riverine flow from upstream, followed by

the post-monsoon, with low values during the

pre-monsoon season (0.5 - 21µmoll-1

, av. 4.1µmoll-1).

The pre-monsoon showed PHc as high as 160 µgl-1,

while the monsoon showed highest PHc value of 102

µgl-1

and the post-monsoon season showed its highest

PHc of 29.3µmoll-1

(Table 1). Trace metals showed highest values of Hg ranging

from 52 – 434 ηgl-1

during the pre-monsoon. During post-monsoon, Hg showed a slight decrease and

ranged from 8 – 216 ηgl-1

, while the monsoon season

showed a further decrease in Hg with values

ranging from 10 – 124 ηgl-1

. The Pb showed higher values at most of the stations during post-monsoon

(1.2 – 4.4 µgl-1

), followed by that during the

pre-monsoon (0.6 – 2.8 µgl-1

), with low values

during monsoon (0 – 1.6 µgl-1

). Cd also showed considerably higher values during the post-monsoon

(0 – 1.3 µgl-1

) and monsoon (0 – 1.1µgl-1

) relative to

pre-monsoon season (0 - 0.3µgl-1

). The Chl. a (1.4 – 19.3 mgm

-3; av. 5.3 mgm

-3) and PP (0.2 -7.6

mgCm-3

; av. 2.7 mgCm-3

) showed higher values

during monsoon and the post-monsoon season (Chl.a: 0.2 – 19.9 mgm

-3; av. 2.75 mgm

-3 and PP: 0 –

19.3 mgCm-3

: av. 2.1 mgCm-3) relative to pre-

monsoon, which showed lowest values (Table 1).

TVC, TC and TV showed higher values during all the 3 seasons, however, the highest values were observed

during the post-monsoon, followed by that during the

pre-monsoon and the least during the monsoon.

Discussion

Estuarine dynamics depend upon tides and riverine inputs. The low and high tide have a large effect on

the amount of nutrients and other water characteristics

that affect the flora and fauna of the estuary. In Zuari estuary, the water from the coastal Arabian Sea enters

more then 10 km upstream of the estuary during high

tide which recedes during the low tide. Majority of

the contaminants entering the estuary along the riverine course are either pushed upwards during the

high tide or downwards during the low tide, and in the

process give rise to increasing contaminants, more particularly, in the estuarine region. This brings about

changes in the water quality of Zuari estuary at

various locations along the estuarine region. One of the approaches for understanding the changes is the

multivariate statistics, wherein factor analysis

(also called Principle Component Analyses – PCA) is

used for identifying the dominant variables. The R-mode (sorted) factor analysis generates the principle

components, which result in eigen values, percentage

of variance and cumulative percentage of the data set,

allowing inter parameter relation and the variation1.

A varimax rotation of different varifactors with

factor loadings calculates the eigen value greater than

1 and are sorted by the results having values greater

than 0.4, based on significant influence25-27

. Rotation of the axis as defined by factor analyses produces a

new set of factors, each one involving primarily a

sub set of the original variables with a little overlap as


317


318

possible so that original variables are divided into

groups. The factor loadings have been classified and

categorized as “strong”, “moderate” and “weak” corresponding to absolute loading values of

0.50 – 0.40 as “weak”, of 0.75 - 0.5 as “moderate”

and of > 0.75 as “strong”28-32

. Data on various water

quality parameters from Mormugao harbour region in Zuari estuary were thus subjected to multivariate

statistics (SPSS 10) using R-mode varimax factor

analyses to understand the dominant variables influencing the quality of harbor water over different

seasons.

R-mode Varimax Factor analyses

Pre-monsoon

High Tide

During the high tide of pre-monsoon, the R-mode

varimax factor analyses of the data indicated a total of

4 factors responsible for 82.7% of the variance (Table 2a). Factor 1 accounted for 37% of the total

variance due to strong positive loadings of TVC,

TC and TV with significant correlations between

them, indicating bacterial dominance in the harbour region. Similarly, the moderate positive loadings of

pH, Chl.a and Hg with good correlations of pH with

Chl.a and Hg suggested the addition of Hg to harbour

water from within the harbour region and moderate

Chl.a generation during the high tide. The dominance

of TVC, TC and TV indicated that they arise from the waste discharges from Mormugao Sewage Treatment

Plant (STP), located closer to the estuarine mouth at

Mormugao Headland, which enter the harbour water

during high tide. Factor 2 explained 61% of cumulative variance and

indicated strong positive loadings of phosphate,

silicates and nitrite, moderate positive loadings of nitrate and BOD and weak positive loading of Pb,

with weak negative loadings of phaeophytin and pH.

The positive correlations of nutrients with each other

and with Pb and BOD and negative with DO suggested a common source and their regeneration from the

oxidation of unoxidised organic matter, mostly from

the STP. The anthropogenic nutrients from the STP thus released supported Chl.a generation in well

oxygenated Mormugao harbour water.

Factor 3 explained 74.5% of cumulative variance and indicated strong positive loading of primary

productivity (PP), moderate positive loading of DO

and moderate negative loading of nitrate. PP

correlated significantly with DO and negatively with nitrate clearly indicating utilization of nitrate and

generation of primary productivity. Factor 4 explained

Table 2 (a,b): Factor loadings of variables during Pre-monsoon.

High tide Pre-monsoon Low tide Pre-monsoon

1 2 3 4 1 2

TVC 0.95 SiO4 0.99

TC 0.93 PP -0.99

TV 0.92 Pb 0.97

PHc -0.84 PHc 0.96

Pb -0.80 0.45 TC -0.96

pH 0.75 -0.45 TVC -0.95

Chla 0.71 TV -0.95

Hg 0.59 NO3 0.95

BOD -0.58 0.55 Hg 0.84 -0.55

PO4 0.88 BOD 1.00

SiO4 -0.50 0.82 PO4 -1.00

NO2 0.82 Chla 0.94

NO3 0.66 -0.61 Cd 0.47 -0.89

PP 0.93 DO 0.53 0.85

DO 0.65 pH 0.60 0.80

Cd 0.83 Pheo 0.66 0.75

Pheo -0.45 0.67 NO2 0.66 -0.75

Eigen 6.3 4.0 2.3 1.4 Eigen 10.0 7.0

% Var. 37.2 23.5 13.7 8.2 % Var. 58.6 41.4

Cumm% 37.2 60.8 74.5 82.7 Cumm% 58.6 100.0

(a) (b)


319

82.7% of cumulative variance and indicated strong

positive loading of Cd and a moderate positive

loading of phaeophytin (deactivated Chl.a), with insignificant correlation between them, suggesting the

dominance of Cd in harbour water.

Low Tide

The factor analyses showed only 2 factors

responsible for 100% of variance in harbour water.

Factor 1 explained 58.6% of the total variance due to

strong positive loadings of silicates, Pb, PHc, nitrate and Hg; moderate positive loadings of nitrite,

phaeophytin, pH and DO with a weak positive loading

of Cd, followed by strong negative loadings of PP, TC, TVC and TV (Table 2b). This showed silicate

dominance in harbour water from riverine input

during low tide. Its significant positive correlation

with nitrate and Pb; that of nitrate with nitrite and phaeophytin; and the positive correlation of Pb with

PHc suggested their contribution to the harbour water

by the riverine water during low tide. The strong negative loadings of PP, TVC, TC and

TV suggested their decrease during the low tide, which

is evident from their observed lower values during the pre-monsoonal low tide relative to high tide.

Factor 2 explained 100% of cumulative variance

and indicated strong positive loadings of BOD, Chl.a,

DO and pH, with moderate positive loading of phaeophytin; strong negative loadings of phosphate,

Cd and nitrite with moderate negative loading of Hg.

The Chl.a correlated well with DO and BOD

and negatively with pH suggesting good Chl.a generation within the harbour water due to well

oxygenation and nutrient availability, evident from

increasing values of Chl.a during the low tide of pre-monsoon relative to high tide. The positive

correlation of BOD with pH indicated its contribution

by Mormugao STP release.

Monsoon

High Tide

A total of 5 factors explained 78.3% of the variance (Table 3a). Factor 1 accounted for 23.8% of the total

variance and indicated moderate positive loading of

DO and weak positive loadings of Chl.a and nitrate, followed by strong negative loadings of phosphate,

lead and silicates and weak negative loadings of PP

and Cd. Chl.a, DO and nitrate do not correlate

significantly with each other, while the weak positive loading of Chl.a suggested low Chl.a generation.

Table 3 (a,b): Factor loadings of variables during Monsoon.

High tide monsoon Low tide monsoon

Variables 1 2 3 4 5 Variables 1 2 3 4 5

pH -0.89 pH -0.68 0.68

DO 0.51 0.72 DO -0.95

BOD 0.78 0.40 BOD 0.74

PO4 -0.90 PO4 0.66 -0.67

NO2 0.95 NO2 -0.68 -0.64

NO3 0.42 -0.42 NO3 -0.40 0.86

SiO4 -0.83 SiO4 0.44 -0.75 0.42

PHc 0.86 PHc 0.46 0.65

Phenol -0.83 Phenol -0.86 0.42

Hg -0.92 Hg -0.92

Cd -0.41 Cd 0.83

Pb -0.87 Pb -0.68

Chla 0.43 0.54 0.55 Chla 0.42 0.49

Pheo 0.57 0.47 Pheo 0.86

PP -0.45 0.63 PP -0.86

TVC 0.77 TVC -0.87

TC 0.70 0.50 TC -0.95

TV -0.86 TV 0.54 0.72

Eigen 4.29 3.68 2.43 1.94 1.75 Eigen 8.40 2.63 2.15 1.54 1.39

% Var. 23.81 20.47 13.50 10.78 9.73 % Var. 46.64 14.62 11.93 8.54 7.73

Cumm% 23.81 44.28 57.78 68.55 78.28 Cumm% 46.64 61.27 73.19 81.74 89.46

(a) (b)


320

Table 1 shows low values of Chl.a from 1.4 to 3.4

mgm-3

during high tide as compared to its high values

from 1.7 to 19.3 mgm-3

during the low tide. Factor 2 accounted for 44.3% of cumulative variance and

indicated strong positive loadings of nitrite and BOD;

moderate positive loadings of Chl.a and phaeophytin;

followed by strong negative loading of pH and a weak negative loading of nitrate. Nitrite correlates

negatively with nitrate but significantly positively

with Chl.a and phaeophytin. The strong negative loading of pH and moderate positive loadings of Chl.a

and phaeophytin suggested that at higher pH, the

Chl.a and phaeophytin decrease, indicating the better

generation of Chl.a and phaeophytin in the brackish estuarine water. Factor 3 accounted for 57.8% of

cumulative variance and indicated moderate positive

loading of DO and weak positive loading of phaeophytin with strong negative loadings of Hg and

phenol. The weak positive loading of phaeophytin

suggested lower degradation of Chl.a in well oxygenated harbour water.

Factor 4 accounted for 68.5% of cumulative

variance and indicated a strong positive loading of

PHc, moderate positive loadings of TC, PP and Chl.a with a weak positive loading of BOD. PHc correlated

positively with all the above mentioned parameters,

while TC correlated positively with BOD, indicating TC to be the contribution from Mormugao STP waste.

The strong positive loading of PHc indicated PHc

contamination from the washings of oil from land, berthing areas in Ports and the river banks. Factor 5

accounted for 78.3% of cumulative variance and

indicated strong positive loading of TVC, moderate

positive loading of TC and strong negative loading of TV. The TVC and TC correlated significantly with

each other but negatively with TV, indicating the

dominance of TVC and TC in harbour water.

Low Tide

A total of 5 factors explained 89.5% of the total variance (Table 3b). Factor 1 accounted for 46.64% of

the variance and indicated a moderate positive loading

of phosphate, weak positive loadings of silicate and PHc, with strong negative loading of

DO and primary productivity, moderate negative

loading of pH and weak negative loading of nitrate.

Phosphate correlated negatively with DO (and pH) and positively with silicates, while silicates correlated

negatively with pH, indicating the phosphate and

silicates as riverine contributions. Similarly, positive correlations of PHc with pH and DO, indicated

addition of PHc to the harbour water by riverine flow

during monsoonal low tide. Factor 2 accounted for

61.27% of cumulative variance and indicated moderate positive loading of TV, weak positive

loading of Chl.a, strong negative loading of TVC and

moderate negative loading of nitrite. Chl.a did not

show any significant relationship with TV, indicating its generation irrespective of the presence of TV in

harbour water. Factor 3 accounted for 73.2% of

cumulative variation and indicated strong positive loading of phaeophytin and nitrate, moderate positive

loading of pH and weak positive loading of Chl.a,

followed by strong negative loadings of Hg, phenol,

moderate negative loadings of silicate, Pb, phosphate and nitrite. Chl.a correlated positively with

phaeophytin, nitrate and pH indicating Chl.a

generation in the harbour water rich in nitrates and its degradation to phaeophytin to be mediated by other

factors (eg. phenol, Hg and Pb).

Factor 4 accounted for 81.74% of cumulative variance and indicated moderate positive loading of TV

and strong negative loading of TC. Both correlated

negatively with each other, indicating the presence of

TV in harbour water and its contribution by Mormugao STP. Factor 5 accounted for 89.5% of cumulative

variance and indicated strong positive loading of Cd,

moderate positive loadings of BOD and PHc with weak positive loadings of silicates and phenol. All these

parameters correlated significantly with each other

indicating their coexistence in harbour water. However, the strong positive loading of Cd indicated Cd

contamination in Mormugao harbour water.

Post-monsoon

High Tide

The factor analyses indicated 4 factors explaining 72% of the total variance (Table 4a). Factor 1 explained 22% of the variance and indicated strong positive loadings of phosphate, PP, silicates and

Chl.a; moderate positive loading of pH and moderate negative loading of Hg. All these parameters do not correlate significantly with each other but the relationship of phosphate with silicate is negative. Silicates are contributed by the riverine water and this relationship indicated that phosphate in harbour water is contributed not by riverine water but from within the harbour water during the post-monsoonal high tide. Similarly, a significant negative relationship of Chl.a with phosphate indicated the generation of Chl.a in the riverine water and their transportation to the harbour water. Moreover, the strong positive loadings


321

of PP and Chl.a in this factor suggested good productivity in harbour water. Table 1 shows an increase in Chl.a from 0.2 to 19.9 mgm

-1 and PP from

0 – 8.8 mgCm-3

. Factor 2 explained 40.6% of cumulative variance and indicated strong positive loading of TV, moderate positive loadings of DO, salinity and temperature, with weak positive loadings of BOD, Hg and phenol followed by moderate negative loading of phaeophytin and weak negative loading of TVC. The strong positive loading of TV indicated its dominance in harbour water, however, the insignificant

positive relationship of TV with salinity and negative with pH, DO and BOD suggested no contribution of TV by riverine water. Moreover, the positive correlation of TV with phenol, indicated the association of phenol with Mormugao STP discharges.

Factor 3 explained 57% of cumulative variance and indicated strong positive loading of TVC, moderate positive loading of phenol and DO, weak positive loadings of Cd and TC followed by a strong negative loading of nitrite and moderate negative loading of nitrate. The strong positive loading of TVC indicated its dominance in harbour water, along with TC. The

insignificant correlation of Phenol with TVC showed

that phenol is not contributed by STP but its presence in well oxygenated harbour water is due to other sources. Factor 4 explained 72% of cumulative variance and indicated strong positive loading of PHc, moderate positive loadings of Pb and TC with moderate negative loadings of BOD and Cd. There is no correlation among the parameters, however, this factor indicated the dominance of PHc in harbour water from boat traffic and shipping activities in the harbour region.

Low Tide

Altogether 4 factors explained 92.6% of the total

variance (Table 4b). Factor 1 explained 33% of the variance and indicated strong positive loadings of

nitrate, phaeophytin and Hg; moderate positive

loadings of salinity and Chl.a with strong negative

loadings of PHc, Pb and temperature followed by moderate negative loading of TV. The strong positive

loading of nitrate indicated its dominance in harbour

water, while its insignificant correlation with Hg and salinity indicated that it is contributed by the estuarine

region. Moreover, the strong negative correlations of

salinity with Chl.a, phaeophytin and Hg, indicated

Table 4 (a,b): Factor loadings of variables during Post-monsoon.

High tide Post Monsoon Low tide Post Monsoon

1 2 3 4 1 2 3 4

PO4 0.88 NO3 0.97

PP 0.85 PHc -0.92

SiO4 0.85 Pheo 0.87

Chla 0.84 Pb -0.85

pH 0.73 Hg 0.82 0.40

Hg -0.59 0.47 Temp -0.82 -0.44

TV 0.87 Salinity 0.66 -0.64

DO 0.72 0.50 TC 0.98

Salinity 0.68 TVC 0.86

Temp. 0.65 NO2 -0.84

Pheo -0.60 PO4 -0.76 -0.54

NO2 -0.85 Chla 0.59 0.68

TVC -0.43 0.83 pH 0.88

Phenol 0.47 0.65 PP 0.43 0.86

NO3 -0.59 TV -0.61 0.69

PHc 0.89 DO -0.44 0.65 0.52

Pb 0.70 SiO4 0.95

BOD 0.49 -0.70 Cd -0.91

TC 0.43 0.612\ BOD 0.58 0.58

Cd 0.45 -0.61 Eigen 6.287 4.312 3.859 3.144

Eigen 4.396 3.731 3.261 2.997 % Var. 33.09 22.7 20.31 16.55

% Var. 21.98 18.654 16.31 14.985 Cumm% 33.09 55.78 76.09 92.64

Cumm% 21.98 40.635 56.94 71.925

(a) (b)


322

their contributions by the riverine water. Factor 2

explained 56% of cumulative variance and indicated

strong positive loadings of TC and TVC; moderate positive loading of Chl.a and weak positive loadings of

PP and Hg followed by strong negative loadings of

nitrite and phosphate with a weak negative loading of

DO. The TVC and TC correlated positively with each other indicating their association, which are contributed

by Mormugao STP, while positive correlations of TVC

with Chl.a, phaeophytin and PP suggested their entry into the harbour water during low tide. This factor thus

indicated the dominance of TVC and TC with good

Chl.a and PP into the harbour water.

Factor 3 explained 76% of cumulative variance and indicated strong positive loadings of pH and PP;

moderate positive loadings of TV, DO and BOD with

moderate negative loading of phosphate and weak negative loading of temperature. No significant

correlations exists among the parameters, except for

positive correlations of PP and pH with DO, suggesting that the harbour water being influenced by

high oxygenated riverine water acts as a better

productive zone and sustains good PP. Factor 4

explained 92.6% of cumulative variance and indicated strong positive loading of silicate; moderate positive

loadings of BOD and DO with strong negative

loading of Cd and moderate negative loading of salinity. BOD and DO correlated positively with each

other but negatively with salinity. Similarly, silicates

correlated negatively with salinity, suggesting that the silicates are brought into the harbour water by the

well oxygenated riverine water during low tide with

some oxidisable organic load. Water Quality Index (WQI)

Water quality index is a mathematical tool, which

transforms the bulk of water quality data into a single digit, cumulatively derived numerical expression,

indicating the level of water quality. It is normally

used for measuring the quality of riverine water and is essential for monitoring the changes in water quality

of a given source as a function of time and other

influencing factors33

. The measured WQI values of parameters of a

sample were then transformed into a single number

called the Overall Index of Pollution (OIP),

representing the overall quality of water at that particular station. The OIP was thus the average of all

the pollution indices (Pi) for individual water quality

parameters considered in the study and is given by the mathematical expression OIP = �i Pi / n where Pi =

pollution index for ith parameter, where, i = 1, 2, . . ., n

and n = number of parameters.

Based on the measured OIP values, the Mormugao harbour water at 8 selected locations S1 to S8 in Zuary

estuary was categorized as per the stipulated OIP

values as shown below.

OIP values and the classification of water.

OIP Values Type of water

0 – 1 Excellent

1 – 2 Acceptable

2 - 4 Slightly polluted

4 - 8 Polluted

8 - 16 Heavily polluted

As per the scheme, the water Quality Index (WQI)

values measured from 8 harbour stations (Table 5) during various seasons showed lowest OIP values for

high tide and low tide of the monsoon season as

compared to other two seasons. Of the remaining two

seasons, the post-monsoon season showed the OIP values in the range of 2 to 4.23 during high tide and

from 1.75 to 5.40 during low tide, indicating

increasing contaminants in harbour water during the post-monsoon season. This increase was greater

during the low tide as compared to that during the

high tide as during this tide, the contaminants from the upstream were brought into the harbour water,

giving rise to polluted water at many locations.

Water Quality Modeling for Assimilative Capacity

The Ecolab module of MIKE 21 was used to

understand the water quality and assimilative capacity

of harbour waters using DO and BOD variation in water. The Ecolab module was validated with

measured DO and BOD values and the model

could reproduce all the features of observations

(Figure 3 a-b). The seasonal variations in water quality was

studied by running the model for the months of

September, December, February and April with effluent load discharges of Mormugao STP (monthly

av. of 380 to 450 m3) and Baina STP (monthly av. of

7 × 106l/d) introduced into the model domain.

Table 5—Tide dependent seasonal variation in OIP values of water in Mormugao harbour region.

Seasons Tide OIP Minimum OIP Maximum

High Tide 1.41 3.52 Pre-monsoon

Low Tide 1.52 1.82

High Tide 1.13 1.69 Monsoon

Low Tide 1.49 1.79

High Tide 2.00 4.23 Post-monsoon

Low Tide 1.75 5.40


323

The DO and BOD concentrations were extracted at

7 different locations, with 4 locations near the mouth region under the influence of Mormugao STP

(Figure 4a) and 3 locations south of the estuary, under

the influence of Baina STP (Figure 4b). The DO and BOD concentrations were also extracted for joint

influence of both the discharges with respect to tidal

variations (Figure 4c). The model indicated no

variation in BOD and DO in all the months within the estuary due to Mormugao STP discharge, the

variation, however, was observed downstream of

the estuary due to Baina STP waste discharge (Figure 5 and Figure 6).

The assimilative capacity of water was estimated

by running the Ecolab module for higher effluent loads introduced into the model at these locations in

the estuary and the DO depletion with respect to BOD

variation was estimated. The results showed that the DO concentration within the estuary for effluent loads

equivalent to Mormugao STP or Baina STP discharge

do not go below 3 mg/l. The estimated DO levels at 4

selected locations for the discharges of 5, 10 and 20 m

3/s are given in Figure 7 and Figure 8.

The model showed that during April, the

load assimilative capacity of the estuarine water was higher than the other three months, as the effluent

Fig. 3—Comparison between measured and modeled DO and BOD concentrations: (a) DO and (b) BOD.

Fig. 4—BOD and DO variations during different months due to Mormugao STP discharge, Baina discharge and both the effluents jointly.


324

Fig. 5—BOD variations with respect to tide elevations within the estuary (a) September 2003, (b) December 2003, (c) February 2004 and (d) April 2004.

discharge up to 10m3/s (864 × 10

6 l/day) can maintain

a DO level above 3mg/l in harbour water, whereas during the other three months, the DO concentration

goes below 3mg/l when the discharge rate is higher

than 5 m3/s (432 × 10

6 l/day). This indicated the role

of physical processes (seasonal effects) controlling

the assimilative capacity.

Within the estuary, there is an increase in BOD

with a marginal variation in DO during the ebb. This may be the result of the retrieval of water

from upstream of the estuary due to tidal effect, which

carries non-point source pollutants into the estuary.

Mixing of water due to wind forcing (winds are of the order of 4 m/s in the afternoon hours) and tidal forcing

increases the aeration of the estuary and maintains well

oxygenated water in the estuary.

Hydrodynamics of Zuari estuary

For a correct assessment of the circulation in an

estuary it is important to understand how the tides act in the estuary and how they can be represented in the

balance of forces.


325

Model domain

To simulate currents in Zuari estuary, the

MIKE 21 hydrodynamic (HD) model was set up using tidal elevations along open boundaries.

The model domain considered extended from

Karwar in the south to Vengurla in the north, with larger and smaller domains within 200 × 200m and

50 × 50m grid size respectively. Depth values were

provided at each grid point. Rectified image of hydrographic map was brought into the background

of a MIKE bathymetry module and the depth

contours were digitized, interpolated and then

exported so as to get the bathymetry of the region.

Using the tidal constituents available for Karwar

and Vengurla region, tides at these two stations

were predicted. The hydrodynamic model was run initially for a larger domain and then the tides for

the open boundaries of the smaller region were

extracted. The simulated tides for Mormugao were compared with the predicted tides (Figure 9) to

ascertain the accuracy of model simulation.

The simulated currents were compared with current meter measurements and the match was found

to be good for both zonal and meridional current

Fig. 6—DO variations with respect to tide elevations within the estuary during (a) September 2003, (b) December 2003, (c) February 2004 and (d) April 2004.


326

Fig. 7—Assimilative capacity of Zuari estuary at a location inside the estuary estimated using effluent discharge of Mormugao STP for (a) September, (b) December, (c) February and (d) April.

components (Fig. 10a-h). Results showed that the

model can reproduce almost all features of the observed currents.

Ecolab is a numerical lab for Ecological Modeling,

which helps in customizing aquatic ecosystem models to describe water quality. The Ecolab module takes care of

the changes in water quality variables occurring due to

physical and biological processes in the system. The

following DO balance equation was used in the model;

dDO/dt = K2 (Cs – DO) – Kd3 BOD d θd3 (T - 20) – K s3

BOD sθs3 (T - 20) – Kb3 BOD b θb3 (T - 20) – Y1K4 NH3

θ4 (T - 20) – R20θ2 (T - 20) + P – B1

The subscripts d, s and b denote dissolved, suspended or settled. For the analyses of a simple BOD-DO model, the Ecolab model is considering only the variation in BOD and DO and hence the equation can be transformed to;


327

Fig. 8—Assimilative capacity of the Zuari estuary at a location inside the estuary estimated using effluent discharge of Baina for (a) September (b) December (c) February and (d) April.


328

dDO/dt = K2 (Cs – DO) – Kd3 BOD d θd3 (T - 20) – K s3

BOD sθ s3 (T - 20) –– R20θ 2 (T - 20) + P – B1

where, B1 - Sediment oxygen demand Cs - Saturation concentration of DO K2 - Re-aeration constant K3 - BOD decay rate K4 - BOD decay rate for nitrification NH3 - Ammonia P - Photosynthesis R20 - Respiration T - Temperature Y1 - Yield factor for DO used for nitrification h2 - Temperature co-efficient for respiration h3 - Temperature coefficient for BOD decay h4 - Temperature coefficient for nitrification

Using this equation, the model estimated the oxygen

balance as the DO depletion was directly related to

BOD in the water column.

To study the changes in water quality due to DO and BOD variation, the effluent discharges from each of these sources was separately introduced into the model and it was run for 7 selected stations, with 4 stations within and around the harbour mouth, assuming that the waste from Mormugao STP affects these stations and 3 stations south of the harbour

mouth (as shown in Fig. 2), assuming that the Baina Municipal waste affects these stations. The quantities and characteristics of effluents released from these two locations were provided by the Goa State Pollution Control Board, Panaji, Goa (India). The model was run for September, December, January, February and April to study the seasonal variations in water quality arising out of the hydrodynamic forcing of the region. To estimate the assimilative capacity of the estuary, the model was run for higher effluent loads than those from the source points and the DO concentrations were

estimated at 4 different locations near the estuarine mouth. A good DO concentration of 3 ppm in water was considered optimum for the biota in the estuary.

The maximum tide elevation inside the Zuari

estuary is of the order of 2.0 m. The simulated ebb and flood currents during December 2003, January

Fig. 10—Comparison between measured and modeled current speeds: (a) December 2003, (c) January 2004, (e) February 2004, (g) April 2004 and current directions: (b) December 2003, (d) January 2004, (f) February 2004, (h) April 2004

Fig. 9—Tides at Mormugao extracted from the larger model domain and predicted using tidal constituents.


329

2004, February 2004 and April 2004 are shown in

Fig. 10(a-h). The flow pattern, when analyzed,

revealed that the water flows into the estuary from the north and flushes out of the estuary primarily towards

south (schematically shown in Fig.2). The flow is

stronger inside the estuary than south of the estuary

(along the coast). The current propagation within the estuary is, in general east-west, mostly influenced by

the tide. At the mouth of the estuary, the northward

flow takes a cyclonic reversal and flows again southwards without entering the estuary.

The northward flow of water entering the estuary

being stronger inside the estuary, drives the

contaminants out of the estuary towards south posing no threat to the water within the harbour area as can

be seen from the factor loadings of low tide.

Conclusion The dominant physico-chemical parameters in

water, seasonal variations in water quality,

assimilative capacity and the hydrodynamic behaviour

of Mormugao harbour water have been studied using R-mode factor analyses, water quality index and the

numerical modeling using MIKE 21 model. The water

showed bacterial dominance of Total Viable Count (TVC), Total Coliforms (TC) and Total Vibrios (TV)

in harbour water during all the seasons. The

deterioration in water quality of the harbour region was less during monsoon while it increased during the

low tide of post-monsoon season due to incoming of

anthropogenic contaminants in the harbour region

from riverine upstream. Of the various organic and inorganic contaminants received by the harbour

water, the presence of nutrients supported C

hl.a generation increasing primary productivity during the post-monsoon.

Water quality index (WQI) studies showed lowest

values during high tide and low tide of the monsoon

season as compared to the other two seasons. Of the remaining two seasons, the post-monsoon season

showed highest WQI values ranging up to 20; with

OIP lying in the range of 2 to 4.23 during high tide and from 1.75 to 5.40 during low tide, indicating

increasing contaminations deteriorating the water

quality of harbour water during post-monsoon. This increase was greater during low tide relative to

high tide as during low tide, the contaminants from

the riverine upstream are brought into the harbour

region, giving rise to slightly polluted water at some stations and polluted water at other stations.

Ecolab module (MIKE 21) suggested higher

assimilative capacity of harbour water during April

2004, with DO levels remaining above 3 mg/l for the effluent discharge up to 864 × 10

6 l/day (10 m

3/s).

The reduction in assimilative capacity was observed

during December, January and February when the DO

level goes below 3 mg/l for effluent discharges of 432 × 10

6 l/day (5 m

3/s). This was attributed to physical

processes controlling the assimilative capacity.

Hydrodynamic module indicated that besides the east-west flowing tidal currents in Zuari estuary, the

coastal water flows into the estuary from the north

and flushes out of the estuary towards south, with a

stronger force inside the estuary. At the mouth of the estuary, the northward flow takes a cyclonic reversal

and flows again southwards without entering the

estuary. This result in flushing of the contaminants out of the harbour area (estuary) posing no much

threat to the quality of Mormugao harbour water.

Acknowledgements

Authors are grateful to Director, NIO, Goa for

providing facilities to carry out the work and to

Mr. K. Sudheesh for his help in using MIKE 21 model.

References 1 Shirodkar P.V., Mesquita, A., Pradhan, U. K., Verlekar, X.

N., Babu, M. T., Vethamony, P., Factors controlling the

physico-chemical characteristics in the coastal waters off Mangalore – A multivariate approach, Environ. Res., 109 (2009) 245-257.

2 Pradhan, U. K., Shirodkar, P. V., Sahu, B. K., Physico-chemical characteristics of the coastal water off Devi estuary, Orissa and evaluation of its seasonal changes using Chemometric techniques, Curr. Sci., 96 (2009) 1203-1209.

3 Dalal, S. G., Shirodkar, P. V., Verlekar, X. N., Jagtap, T. G.,

Rao, G. S., Apportionment of sources affecting water quality: Case study of Kandla Creek, Gulf of Kutchchh, Environ.

Forensics., 10 (2009) 103-108. 4 Ujevic, I., Odzak, N., Baric, A., Trace metal accumulation in

different grain size fractions of sediments from a semi enclosed Bay, heavily contaminated by urban and industrial waste waters, Water Res., 34 (2000) 3055-3061.

5 Carrasco, M., Lopez Ramirez, J.A., Benvente, J., Lopez

Aguayo, F., Sales. D., Assessment of urban and industrial contamination levels in the Bay of Cadiz, SW Spain, Mar.

Pollut. Bull., 46 (2003) 335-345. 6 Prego, R., Cobelo- Garcia, A., Twentieth Century of

overview of heavy metals in the Galician Rias (NW Iberian Peninsula), Environ. Pollut., 121 (2003) 425-452.

7 Wollast, R., Wefer, G., Lamy F., Mantoura, F., Biogeochemical processes in estuaries. (Eds.), Marine

Science Frontiers for Europe. (Springer-Verlag, Berlin) 2003 pp 61-77.


330

8 Owen, R. B., Sandhu, N., Heavy metal accumulation and anthropogenic impacts on Tolo Harbour, Hong Kong, Mar.

Pollut. Bull., 40 (2000) 174-180. 9 Thomann, R., Mueller, J., Principle of surface water quality

modeling and control, (Harper and Row publishers New York) 1987.

10 Hoppensteadt, F.C., Mathematical methods of Population

Biology, (Cambridge University Press) Cambridge. 1982. 11 Jorgensen, S.E., Fundamentals of Ecological Modelling, (3rd

ed. Elsevier, Amsterdam) 2001.

12 Gertsev, V. I., Getrseva, V. V., Classification of mathematical models in ecology, Ecol. Model., 178 (2004) 329–334.

13 Tyagi, B., Gakkar, S., Bhargava. D. S., Mathematical modeling of stream DO-BOD accounting for settle able BOD and periodically varying BOD source, Environ. Model.

Softw., 14 (1999) 461–471. 14 Gupta, I., Dhage, S., Chandorkar, A.A., Srivastav. A.,

Numerical modeling for Thane Creek, Environ. Model.

Softw., 19 (2003) 571–579. 15 Desa, E., Zingde, M. D., Vethamony, P., Babu M. T.,

D’Souza, S. N., Verlecar. X. N., Dissolved oxygen – a target indicator in determining the use of the Gulf of Kutchchh waters, Mar. Pollut. Bull., 50 (2004) 73-79.

16 Vethamony, P., Reddy, G. S., Babu, M. T., Sudheesh, K., Tidal eddies in a semi enclosed basin: a model study, Mar.

Environ. Res., 59 (2005) 519-532. 17 Strickland, S.C., Parsons, T.R., A practical handbook of

seawater analyses, (Bulletin of Fisheries Research Board of Canada, Ottawa) 1979.

18 APHA, Standard methods of seawater analyses, (American Public Health Association, New York) 1992.

19 Grasshoff, K., Methods of seawater analyses, (Verlag-Chemie, Weinheim), 1999.

20 Anonymous, MIKE 21 user guide and reference manual,

(Horshoim, Denmark; DHI, Danish Hydraulic Institute) 2001.

21 Chubarenko, I., Tehepikova, I., Modelling of man-made contribution to salinity increases into the Vistula lagoon (Baltic Sea), Ecol. Model., 138 (2001) 87-100.

22 Madsen, H., Jacobson, F., Cyclone induced storm surge and

flood forecasting in the northern bay of Bengal, Coast. Eng., 51 (2004) 277-296.

23 Babu, M. T., Vethamony, P., Desa, E., Modeling tide-driven currents and residual eddies in the gulf of Kutchchh and their

seasonal variability: a marine environmental planning perspective, Ecol. Model., 184 (2005) 299-312.

24 Babu, M. T., Kesava Das, V., Vethamony, P., BOD-DO modeling and water quality analysis of a waste water outfall off Kochi, west coast of India, Environ. Int., 32 (2006) 299-

312.

25 Sahu, K. C., Panda, U. C., Mohapatra, D. M., Geo-chemistry and mineralogy of sediments in Rushikulya estuary, east coast of India, Chem. Environ. Res., 7 (1998) 77–92.

26 Panigrahy, P. K., Das, J., Sahoo, R. K., Evaluation of the influence of various physico-chemical parameters on coastal water quality around Orissa by factor analysis, Indian J. of

Mar. Sci., 28 (1999) 360-364.

27 Rath, P., Bhatta, D., Sahoo, B. N., Panda, U. C., Multivariate

statistical techniques in hydro-geochemical studies: an example from Karnataka, India, Wat. Res., 36 (2000) 2437-2442.

28 Liu, C. W., Lin, K. H., Kuo, Y. M., Application of factor analysis in the assessment of ground water quality in a black

foot disease area in Taiwan, Sci. Total Environ., 313 (2003) 77–89.

29 Vega, M., Pardo, R., Barrado, E., Deban, L., Assessment of seasonal and polluting effects on the quality of river water by exploratory data analysis. Wat. Res., 32 (1998) 3581-3592.

30 Helena, B., Pardo, R., Vega, M., Barrado, E., Fernandez, J.M., Fernandez, L., Temporal evolution of ground water composition in alluvial (Pisuerga River, Spain) by principal component analysis, Wat. Res., 34 (2000) 807–816.

31 Simenova, P., Simenova, V., Andrew, G., Analysis of the Struma river water quality, Cent. Eur. J. Chem., 2 (2003) 121–126.

32 Singh, K. P., Malik, A., Mohan, D., Sinha, S., Multivariate statistical techniques for the evaluation of spatial and

temporal variations in water quality of Gomati River (India) – A case study, Wat. Res. 38 (2004) 3980-3992.

33 Sargaonkar, A., Deshpande, V., Development of an overall Index of Pollution for surface water based on a general classification Scheme in Indian Context. Environ. Monit.

Assess., 89 (2003) 43-67.

Tide dependent seasonal changes in water quality and assimilative capacity of anthropogenic influenced Mormugao harbour water

Documents